HOMOGENEITY TEST FOR CORRELATED DATA IN OPHTHALMOLOGIC STUDIES


Changxing Ma, Kejia Wang

Introduction



       In ophthalmologic studies, measurements obtained from both eyes of an individual are often highly correlated. Ignoring the correlation could lead to incorrect inferences. Rosner (1982) proposed a parametric model and a test statistic for testing homogeneity of proportions among g groups accounting for inter-class correlation, however, the maximum likelihood estimates (MLEs) and likelihood-based tests were not given. Ma and Shan (2013) derived a efficient algorithm for MLE under Rosner's model and theree testing procedures for this problem. This calculator provides a user-friendly interface to use this methodology.
      
Additional simulation results

Calculator


■ DATA


       Please put your data in the box below,
or try the example 1, or the example 2 in the paper,
or simulate data - turn simulation panel On Off
3 × g   matrix for bilateral patients
followed by 2 × g   matrix for unilateral patients


■ OPTIONS


       - Dispay Details of Iterations
       - Termination tolerance on estimates, a positive scalar:

■